Bill's Bad Advice

 I sometimes give the following advice for research which I label Bill's Bad Advice. We will later see who it might be good advice for. Spoiler alert: the number of people for whom it is good advice is shrinking but might include Lance especially now (see his post about stepping down from admin, here).

When you come across a possible research topic or problem, or have some idea,  and are wondering if you want to pursue it,  here is my bad advice: 

1) DON"T  CARE if anyone else cares. If YOU care then that is enough to at least get started. 

2) DON"T  CARE if it has the potential for a published paper. FIRST do the work then, if you feel like it, look for a good venue. You might not bother if posting to arxiv or making an open problems column out of it or a (guest) blog post out of it is  a good endpoint.  (I run the SIGACT News Open Problems Column- feel free to contact me if you want to submit one.) 

3) DON"T CARE if it has practical implications. 

4) DON"T  CARE if you can get a grant for it. With the current state of the NSF this advice may soon become irrelevant. 

5) DON"T  CARE if someone else already did it (though at a later stage you should check on this). Even if you work on it and find someone else did it, you will have LEARNED about the problem through your efforts. You might then want to do a survey for your own benefit to consolidate your  knowledge. 

Why should you NOT CARE about any of these things? Because they get in the way of actually DOING something.  

Here are two examples of when this approach WORKED and one where it DID NOT WORK, though both classifications might depend on your definition of WORKED. 

WORKED: My work on Muffins. All I wanted was to get some High School and Under Grad Projects out of it. I ended up with a book on it which made me twenty dollars last year! More to the point, I learned a lot, as did my co-authors and I am happy with the book. The co-authors were undergraduates so my dept put me up for a mentoring award (I have other credentials as well). I did not win, but I got a nice letter saying they had many qualified applicants. OH- it didn't say if I was one of them. 

WORKED: I have had many Ramsey Projects where a High School Student codes stuff up and learns some Ramsey, some coding, and gets the experience of research. Sometimes they do a survey paper or open problems column. We both learn A LOT from this and the student gets a good letter from me. Do they do something NEW? Publishable? No, though some survey's and open problems columns have come out of this.  I DO tell them ahead of time that the work is unlikely to lead to original results (and hence unlikely to be good for a science competition).   

DID NOT WORK: See this blog post here about the math and here about finding out that the problem we were working on was already known and more was known than we thought. I didn't mind this, but one of the authors did. 

Note that WORKED and DID NOT WORK also depend on your goals. 

For whom is this bad advice? Good advice? 

1) It was always bad advice for young assistant professors who need to get papers and grants to get tenure. 

2) Hypothetically, once you get tenure you have job security and hence can change fields or follow my bad advice without consequences. But with grants and salary and teaching load issues, this is less the case. Perhaps I am nostalgic for a time that never was.

3) High School Students were my main audience for bad advice. Its not as important for them to get papers out as for (say) assistant professors. But even this is changing. Colleges are getting more competitive. And HS students may well want a project that can lead to Science competitions. I am not going to say things were better when I was a kid but instead pose non-rhetorical questions:

a) Are high school students getting into research earlier than they used to? I am sure the answer is yes.

b) Are we losing the safe space where a high school student can just learn things and do things and not worry so much about if its publishable?  Yes, but I am not sure how widespread that is.

c) If we are losing that safe space, is that a bad thing? 

d) Points a,b,c apply to ugrads who want to go to grad school more than for high school students who want to go to college. 

4) Full Professors may have more freedom to follow my bad advice. Lance is looking for things to do now that he is no longer a dean, and indeed, is back to being a teaching-and-research professor. So he might follow my advice. However, he actually cares if people care about his work. He does not have to follow all of my advice, but he can follow some of it.